当前位置: X-MOL 学术Cluster Comput. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Optimal VM-to-user mapping in cloud environment based on sustainable strategy space theory
Cluster Computing ( IF 4.4 ) Pub Date : 2021-06-14 , DOI: 10.1007/s10586-021-03323-2
Pouria Khanzadi , Sepideh Adabi , Babak Majidi , Ali Movaghar

According to the previous studies in the field of economics-oriented cloud resource allocation using game theory, finding an equilibrium point for price is difficult in many cases. This is due to the stochastic situations of the cloud market. So, to tackle this problem we should find a space to describe behavior of the players’ strategies that be independent from the equilibrium point. Therefore, a new algorithm for VM-to-user mapping in the cloud market called VMUMA is proposed. For designing VMUMA, the cloud market is modeled by game theory. VMUMA is based on Sustainable Strategy Space Theory (SSST) in which, each stability and instability of the player’s strategies in the game are defined and all of the interactions in the game are modeled in a two-dimensional space. Then, it is proved that stability and instability of a strategy are quantifiable if minimum information and uncertainty of the game event are determinable. Furthermore, we find a point in the two-dimensional space in which stability and instability of the strategies are equal. Based on this result, it is proved that a strategy is sustainable if this point has minimum variations. Finally, it is proved that there is a transformation for a square matrix in which the matrix is mapped to summation and subtraction of its elements and vice versa. Based on this transformation, it is shown that area of this space is related to the stability and instability of the players’ strategies. The obtained numerical simulation results show that VMUMA can allocate the resources in the market according to the randomness of stability and instability of the strategies.



中文翻译:

基于可持续策略空间理论的云环境下最优VM-to-user映射

根据以往经济学导向的云资源配置领域的博弈论研究,在很多情况下很难找到价格的均衡点。这是由于云市场的随机情况。因此,为了解决这个问题,我们应该找到一个空间来描述独立于均衡点的参与者策略的行为。因此,提出了一种称为VMUMA的云市场中VM到用户映射的新算法。为设计 VMUMA,云市场采用博弈论建模。VMUMA 基于可持续策略空间理论 (SSST),其中定义了游戏中玩家策略的每个稳定性和不稳定性,并且游戏中的所有交互都在二维空间中建模。然后,证明如果博弈事件的最小信息和不确定性是可确定的,则策略的稳定性和不稳定性是可量化的。此外,我们在二维空间中找到了一个点,其中策略的稳定性和不稳定性相等。基于此结果,证明如果该点具有最小变化,则策略是可持续的。最后,证明了方阵存在变换,其中矩阵映射到其元素的求和和减法,反之亦然。基于这种变换,表明该空间的面积与参与者策略的稳定性和不稳定性有关。得到的数值模拟结果表明VMUMA可以根据策略的稳定性和不稳定性的随机性来配置市场中的资源。

更新日期:2021-06-15
down
wechat
bug